[febrianbaehaqi]

Hey guys, I'm new here. And new too for using OpenFOAM. I try to use pimpleFoam as my solver and getting my courant number exploding (and my deltaT are always getting smaller and smaller).

Here is the example of my running command:

Courant Number mean: 0.0227576 max: 14.7084
deltaT = 0.000403384
Time = 0.00159386

PIMPLE: iteration 1
smoothSolver: Solving for Ux, Initial residual = 0.186055, Final residual = 7.95693e-06, No Iterations 2
smoothSolver: Solving for Uy, Initial residual = 0.135143, Final residual = 9.63007e-06, No Iterations 8
smoothSolver: Solving for Uz, Initial residual = 0.107278, Final residual = 9.49794e-06, No Iterations 11
GAMG: Solving for p, Initial residual = 0.04529, Final residual = 0.000394791, No Iterations 5
time step continuity errors : sum local = 2.40771e-06, global = 7.02496e-07, cumulative = -7.93094e-06
GAMG: Solving for p, Initial residual = 0.388235, Final residual = 3.87086e-08, No Iterations 17
time step continuity errors : sum local = 2.17783e-11, global = 1.00988e-11, cumulative = -7.93092e-06
smoothSolver: Solving for epsilon, Initial residual = 4.92616e-05, Final residual = 7.96414e-06, No Iterations 5
bounding epsilon, min: -8822.53 max: 2.9206e+06 average: 36583.1
smoothSolver: Solving for k, Initial residual = 0.000135808, Final residual = 9.87592e-06, No Iterations 7
ExecutionTime = 108.63 s ClockTime = 120 s

Courant Number mean: 0.00788521 max: 10.0964
deltaT = 0.000200025
Time = 0.00179389

PIMPLE: iteration 1
smoothSolver: Solving for Ux, Initial residual = 0.00294686, Final residual = 2.05597e-06, No Iterations 1
smoothSolver: Solving for Uy, Initial residual = 0.074684, Final residual = 8.23839e-06, No Iterations 2
smoothSolver: Solving for Uz, Initial residual = 0.0350147, Final residual = 5.42758e-06, No Iterations 2
GAMG: Solving for p, Initial residual = 0.40826, Final residual = 0.00307529, No Iterations 2
time step continuity errors : sum local = 3.01699e-07, global = -5.50991e-08, cumulative = -7.98602e-06
GAMG: Solving for p, Initial residual = 0.0600293, Final residual = 6.23969e-08, No Iterations 14
time step continuity errors : sum local = 4.70306e-12, global = -2.40713e-12, cumulative = -7.98603e-06
smoothSolver: Solving for epsilon, Initial residual = 1.07313e-05, Final residual = 1.44193e-06, No Iterations 1
bounding epsilon, min: -14486.5 max: 9.9375e+08 average: 62701.3
smoothSolver: Solving for k, Initial residual = 2.39836e-05, Final residual = 3.16133e-06, No Iterations 1
ExecutionTime = 137.2 s ClockTime = 153 s

Courant Number mean: 0.0039057 max: 22.0363
deltaT = 4.53918e-05
Time = 0.00183928

PIMPLE: iteration 1
smoothSolver: Solving for Ux, Initial residual = 0.000530224, Final residual = 3.26394e-08, No Iterations 1
smoothSolver: Solving for Uy, Initial residual = 0.00546058, Final residual = 1.32034e-06, No Iterations 1
smoothSolver: Solving for Uz, Initial residual = 0.00329133, Final residual = 8.3871e-07, No Iterations 1
GAMG: Solving for p, Initial residual = 0.528485, Final residual = 0.00405198, No Iterations 2
time step continuity errors : sum local = 3.05092e-08, global = -7.75921e-10, cumulative = -7.9868e-06
GAMG: Solving for p, Initial residual = 0.0102294, Final residual = 9.89808e-08, No Iterations 11
time step continuity errors : sum local = 1.12747e-12, global = -1.80682e-13, cumulative = -7.9868e-06
smoothSolver: Solving for epsilon, Initial residual = 0.00137331, Final residual = 1.43467e-06, No Iterations 3
bounding epsilon, min: -26025.6 max: 6.31434e+11 average: 930904
smoothSolver: Solving for k, Initial residual = 0.000201095, Final residual = 5.1694e-06, No Iterations 1
ExecutionTime = 162.63 s ClockTime = 182 s

Courant Number mean: 0.000886184 max: 23.5366
deltaT = 9.64372e-06
Time = 0.00184892

PIMPLE: iteration 1
smoothSolver: Solving for Ux, Initial residual = 0.000379375, Final residual = 4.63077e-08, No Iterations 1
smoothSolver: Solving for Uy, Initial residual = 0.00267864, Final residual = 1.78884e-06, No Iterations 1
smoothSolver: Solving for Uz, Initial residual = 0.0017895, Final residual = 8.39166e-07, No Iterations 1
GAMG: Solving for p, Initial residual = 0.587346, Final residual = 0.00437625, No Iterations 2
time step continuity errors : sum local = 5.44944e-09, global = -1.5064e-10, cumulative = -7.98695e-06
GAMG: Solving for p, Initial residual = 0.0126355, Final residual = 6.86878e-08, No Iterations 11
time step continuity errors : sum local = 1.35882e-13, global = -3.0686e-14, cumulative = -7.98695e-06
smoothSolver: Solving for epsilon, Initial residual = 0.013023, Final residual = 1.57433e-06, No Iterations 1
bounding epsilon, min: -3.27318e+10 max: 4.26397e+13 average: 4.06661e+07
smoothSolver: Solving for k, Initial residual = 0.000413744, Final residual = 8.34061e-06, No Iterations 7
ExecutionTime = 188.84 s ClockTime = 212 s

I tried to decrease my time step value but it seems did not work because the problem keep happening. I'm using snappyHexMesh to create the mesh and when I performed checkMesh, the mesh are okay but before the checkMesh end it says "Failed 1 mesh checks". Could someone give me suggestion please? I appreciate that. Thank you.

[Andrea1984]

Hi,

without any other details about your case I would say that from your output it looks like you are using PIMPLE in PISO mode, i.e. with nOuterCorrectors = 1. With these settings your max Co should be < 1.

If you want to make the most out of PIMPLE and use Co > 1 you'll have to have nOuterCorrectors > 1.

See this page for more details on PIMPLE: https://openfoamwiki.net/index.php/O...hm_in_OpenFOAM

Andrea

[febrianbaehaqi]

Hi Andrea, thank you for your answer. I really appreciate that.

I have followed your instruction to add nOuterCorrectors from the link to my fvSolution. But it seems that the problem keeps appearing.

I have tried to fix all my meshes and when I check the mesh using checkMesh command it says that all okay.

Here is the system directory that I am using.

Thank you.

[Andrea1984]

Hi Febrian,

in my (limited) experience with PIMPLE you have to play a little bit with the various parameters to obtain a stable and reasonably accurate solution with Co > 1.

The main parameters I play with are nCorrectors and nOuterCorrectors in fvSolution and maxCo in controlDict. Depending on my mesh orthogonality, I can also throw nNonOrthogonalCorrectors in the mix. I don't see any issue in using a higher number or outer corrector loop ( as nOuterCorrectors =50 in your example) but I usually make sure that my outer correction step converges in a lower number in iterations (~10) once the calculation is stable. In OF v6 the convergence criteria for the outer cycle are specified via the outerCorrectorResidualControl keyword.

Usually the entire process takes a few attempts to find the best combination of the different coefficients. If your case is large, try to set up a simplified case - still including the relevant physics of the original case - and try to find your best combination there.

Also I don't know anything about the physics of your problem, but you might want to use a low maxCo in the beginning of the transient, and then increase it once you reach a more "stable" status, if one such status exists.

Andrea